The Training of Good Physicians: Critical Factors in Career Choices [Reprint 2014 ed.] 9780674420991, 9780674420984


146 70 13MB

English Pages 262 [264] Year 1968

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Acknowledgments
Contents
Tables
I. Why Study Doctors' Career Decisions?
II. Study Methods and Approach
III. Present Practice Circumstances
IV. Decisions about a Field of Practice
V. Hospital Training Decisions
VI. How the Graduates Viewed Their Medical Training
VII. Discussion and Conclusion
References
Index
Recommend Papers

The Training of Good Physicians: Critical Factors in Career Choices [Reprint 2014 ed.]
 9780674420991, 9780674420984

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

The Training of Good Physicians

THE

TRAINING

GOOD

OF

PHYSICIANS

Critical Factors in Career Choices

by Fremont f . Lyden / H. Jack Geiger / Osler L.

Peterson

A C O M M O N W E A L T H F U N D BOOK Harvard University Press, Cambridge, Massachusetts / 1968

© Copyright 1968 by the President and Fellows of Harvard College All rights reserved

Distributed in Great Britain by Oxford University Press, London

Library of Congress Catalog Card Number 68-21977 Printed in the United States of America

Acknowledgments A R E T R O S P E C T I V E S T U D Y of the training and career decisions of nearly two thousand physicians who graduated from twelve widely scattered medical schools required unusual care in planning. W e were particularly fortunate to have the frequent advice and help of Dr. Sol Levine during the preparatory phase of the study reported here. His help was both detailed (as in the construction of the questionnaire) and of a more general character, as when we outlined the steps to be taken to assure a high response rate. Edwin B. Hutchins of the Association of American Medical Colleges assisted us greatly in selecting medical schools for inclusion in this study and also made available the Medical College Admission Test Scores for the physicians in our study population. The extensive statistical work that was necessary to produce the several hundred tables upon which our results are based was organized and completed by Mrs. Mary Wood McDonald. Her meticulous attention to detail and her competent and systematic completion of tables was a vital first step in data handling. Miss Katharine Hendrie, who performed a large number of statistical computations, programmed much of our material for computer treatment, and patiently checked our many tables and figures to assure their accuracy, has been of inestimable help in completing this volume. A special note of thanks is due to the deans and assistant deans of twelve medical schools who aided us by supplying information about the medical school records of their graduates and whose assistance in follow-up of nonresponders contributed very importantly to the overall results and especially to the very

vi

ACKNOWLEDGMENTS

high response rate achieved. Our preoccupation with obtaining a very high response rate led us, possibly unnecessarily, to assure each school that the fact of its participation would be held in confidence. This makes it impossible for us to identify individually the many medical school officers whose help has been so great. This study could not have been done without the cooperation of nearly two thousand practicing doctors. As a group they responded to our questions and demands with commendable promptness and thoroughness. Special mention should be made of the help given us by Professor William G. Cochran of the Faculty of Arts and Sciences, Harvard College, and by Dr. Theodore Colton of the Department of Preventive Medicine of Harvard Medical School. Professor David D. Rutstein has given much advice and encouragement. Miss Jean Haley and Miss Jan North, who have worked extensively on this manuscript, have greater reason than anyone else, perhaps, to take pleasure in seeing it completed. They have allowed us to revise it repeatedly with remarkably good humor. During the period of planning, study execution, and first stages of analysis, Drs. Lyden and Geiger were members of the Department of Preventive Medicine, Harvard Medical School. Dr. Lyden was supported by a fellowship of the Health Information Foundation and Dr. Geiger was a Postdoctoral Research Fellow, Joint Program in Social Science in Medicine of the Department of Social Relations of Harvard University. Support for this study was provided by T h e Rockefeller Foundation.

Contents I

W h y Study Doctors' Career Decisions?

II

Study Methods and Approach

III

Present Practice Circumstances

IV

Decisions about a Field of Practice

V

Hospital Training Decisions

VI

H o w the Graduates Viewed Their Medical Training

VII

Discussion and Conclusion

References Index

243

241

1

6 19 48

104

212

Tables 1

Percentage distribution of doctors by present field of practice, 1961

2

20

Percentage distribution of doctors by amount of training they planned to obtain

3

22

Percentage distribution of general practitioners by amount ot training they planned to obtain

24

4

Percentage distribution of doctors by age

27

5

Percentage distribution by all doctors and doctors in certain fields of practice by age

6

28

Percentage distribution of doctors by present type of practice relationship and type anticipated at peak of career

7

Percentage distribution of general practitioners

30 by

present type of practice relationship and type anticipated at peak of career 8

32

Percentage distribution of internists by present type of practice relationship and type anticipated at peak of career

9

35

Percentage distribution of surgeons by present type of practice relationship and type anticipated at peak of career

10

37

Percentage distribution of doctors by present location of practice

38

TABLES

11

IX

Percentage distribution of all groups by location of present practice

40

Percentage distribution of doctors by type of hospital appointment

41

Percentage distribution of doctors by 1961 taxable income

43

Percentage distribution of doctors reporting salary income as proportion of total medical income

44

Percentage distribution of doctors by value orientation and by field of practice

47

Percent of doctors strongly influenced in choice of a field of practice by factor and field of practice

50

Percentage distribution of doctors by father's education

56

Percentage distribution of employed persons by ocpation and occupations of fathers of medical school graduates

57

19

Percentage distribution of doctors by place of birth

62

20

Percentage distribution of doctors by age at graduation

65

21

Percent of doctors reporting a major source of financial support for medical education by source

67

Percent of doctors reporting minor sources of financial support for medical education by source

70

Percentage distribution of doctors by yearly earnings while in medical school

71

12 13 14 15 16 17 18

22 23

X

24

TABLES

Percentage distribution of doctors by debt at graduation

72

Percent of doctors debt-free at conclusion of medical education

74

26

Percentage distribution of doctors by marital status

78

27

Percentage distribution of doctors by type of internship

82

28

Percentage distribution of doctors by field of practice and type of internship

83

Percentage distribution of doctors by hospital of internship

85

Percentage distribution of doctors by hospital of residency

87

Percent of doctors reporting 24 months or more residency training

88

Percent of doctors with residency and average duration of residency

90

Percent of doctors by field of practice and by composite M C A T score

91

Percentage distribution of doctors by field of practice and class rank

93

Percentage distribution of all doctors by time of final training decision and by field of practice

97

Percentage distribution of all doctors and general practitioners by amount of training decided upon

98

25

29 30 31 32 33 34 35 36

TABLES

37

38

39

40

41

42 43 44 45 46

XI

Percent of doctors reporting encouragement in planning training by people influencing training beyond internship

100

Percent of doctors strongly influenced in choice of field of practice, by influential factor and type of internship

106

Percent of doctors strongly influenced in choice of field of practice by influential factor and by hospital of internship

108

Percent of doctors strongly influenced in choice of field of practice by influential factor and months of residency

109

Percent of doctors strongly influenced in choice of field of practice by influential factor and by class rank in thirds

113

Percentage distribution of doctors by hospital of internship and months of residency

116

Percent of doctors reporting residency by type of internship

117

Percentage distribution of 1954 graduates by M C A T scores

119

Percentage distribution of doctors by type of internship and by composite M C A T score

119

Percent of doctors by hospital of internship by composite M C A T score

120

Xll

TABLES

47

Percent of doctors with no residency by M C A T score

121

48

Percent of doctors by class rank with rotating or straight medical iriternship

123

Percent of doctors reporting major teaching hospital internship by class rank

1-24

50

Percent of doctors with residency training by class rank

124

51

Percentage distribution of doctors by father's education

126

Percentage distribution of doctors by father's education and hospital of internship

127

Percentage distribution of doctors by father's education and by hospital of residency

129

Percentage distribution of doctors of Eastern European and Eastern Mediterranean background by class rank

130

Percentage distribution of doctors by father's occupation and by hospital of internship, sons of craftsmen and all others

133

Percentage distribution of doctors by father's occupation and by hospital of residency, sons of craftsmen and all others

135

Percentage distribution of doctors by father's occupation, by months of residency

136

Percentage distribution of doctors born in large cities or in small towns by class rank

137

49

52 53 54

55

56

57 58

TABLES

59 60 61 62 63 64 65

66 67 68 69 70

Xlll

Percentage distribution of doctors reporting psychiatric residency and all others by place of birth

139

Percentage distribution of doctors by age at graduation and class rank

140

Percentage distribution of doctors by age at graduation and hospital of internship

142

Percentage distribution of doctors with residency and doctors with no residency by age at graduation

144

Percent of doctors by rotating or straight medical or surgical internships by major source of support

147

Percent of all doctors with major teaching hospital or other internships by major source of support

149

Percentage distribution of doctors reporting major support by parents and all others by months of residency

150

Percent of all doctors with major teaching and other hospital residency by major source of support

151

Percentage distribution of doctors by yearly earnings in medical school and by type of internship

153

Percentage distribution of doctors by yearly earnings in medical school by hospital of internship

154

Percentage distribution of doctors by yearly earnings while in medical school by hospital of residency

156

Percentage distribution of doctors by yearly earnings while in medical school by months of residency

157

xiv 71

72

73

74

75

76

77

78

79 80

81

82

TABLES

Percentage distribution of doctors with rotating and straight medical or surgical internships by amount of debt at graduation

159

Percentage distribution of doctors by debt at graduation and hospital of internship

161

Percentage distribution of doctors by residency or no residency and amount of debt at graduation

162

Percentage distribution of doctors by hospital of internship and marriage before entering medical school

166

Percentage distribution of doctors by marriage before medical school, by months of residency

167

Percentage distribution of doctors by M C A T score and class esteem ranking

171

Percentage distribution of doctors by class rank and class esteem ranking

172

Percentage distribution of class esteem leaders and other students by type of internship

172

Percent of doctors by class esteem ranking with major teaching hospital training

173

Percent of doctors by class esteem ranking with no residency

173

Percentage distribution of doctors by friendship group and by hospital of internship

174

Percent of doctors who were encouraged to obtain residency training by source of encouragement and by type of internship

178

T A B L E S

83

84

85

86

87

88

89

90

91

92

XV

Percent of doctors who were encouraged to obtain residency training by source of encouragement and by hospital of internship

180

Percent of doctors who were encouraged to obtain residency training by source of encouragement and hospital of residency

182

Percent of doctors who were encouraged to obtain residency by source of encouragement and by months of residency

183

Percent of doctors who were encouraged to obtain residency by source of encouragement and by class rank

184

Percent of doctors who were encouraged to obtain residency by source of encouragement and M C A T scores

185

Percentage distribution of doctors by dissatisfaction with medical education

186

Percentage distribution of doctors by dissatisfaction with medical education, general practitioners and all others

188

Percentage distribution of doctors by composite M C A T scores and dissatisfaction with medical education

189

Percentage distribution of doctors by hospital of internship and dissatisfaction with medical education

190

Percent of doctors with residency and no residency by dissatisfaction with medical education

192

XVI

93 94

95 96 97 98 99 100 101 102

103 104

TABLES

Percent of doctors dissatisfied with medical education, by months of residency

193

Percentage distribution of all doctors and general practitioners by major criticism of medical education

194

Percentage distribution of doctors stating criticism by composite M C A T scores and by major criticism

196

Percent of doctors dissatisfied with clinical training by academic rank

197

Percent of general practitioners and all other doctors dissatisfied with hospital training

198

Percentage distribution of doctors by opinion about hospital training by M C A T score

199

Percent of doctors with residency or no residency who were dissatisfied with hospital training

199

Percent of doctors dissatisfied with hospital training by months of residency

200

Percentage distribution of doctors reporting limitations of training by major responsible factor

201

Percentage distribution of general practitioners and all other doctors by frequency of curtailment of training, by responsible factor

203

Percentage distribution of doctors choosing the same or different training

205

Percent of doctors preferring the same internship as received by field of practice

206

TABLES 105 106

107 108

xvii

Percentage distribution of doctors by type of internship received and type preferred

207

Percentage distribution of doctors by preferred residency

208

Percentage distribution of doctors by preferred residency and by field of practice

209

Percentage distribution of doctors by preferred residency and by months of residency

210

I / W h y Study Doctors' Career Decisions? E V I D E N C E indicates that most of the college students who intend to go to medical school ultimately obtain entry. These young people are from many diverse family, social, and educational backgrounds; when they enter medical school, they are forced into an educational program which, within each institution, is as uniform for all students as any program can be. THE

W h e n these students leave medical school, they immediately pursue diverse programs of training. The internship and residency training programs they follow vary both in quality and length. Similarly, when they subsequently enter practice, they select many different and varied careers; these differences include place of practice, specialization or field, and the economic and professional organization of practice. Unfortunately, there then appears still another kind of variation: in clinical skills, or—to characterize it more broadly—in the quality of practice. Numerous studies (and every physician's or medical educator's personal experience) demonstrate that doctors do vary in knowledge, skill, and the quality of their work. The several studies of general practitioners or other family doctors have all concluded that doctors' clinical skills are related to the amount of their clinical training, and suggest that, in general, longer hospital training is preferable to short training and that teaching-hospital experience is more effective preparation for practice than nonteaching-hospital experience. (6, 22, 23) In short, then, a rather diverse group of students who undergo a very uniform medical school experience nevertheless maintain their diversity in their choice of further training and careers.

2

THE

TRAINING

OF

GOOD

PHYSICIANS

One reason for the importance of these choices is that they may have some influence on the quality of subsequent work. T h e present study is not concerned with whether or not doctors achieve clinical distinction. It is concerned with the differences between the doctors who prepare "well" for practice and those who do not. As part of this, it is concerned with the characteristics of the doctor which may influence him to obtain a "good" hospital training or that may influence him to enter practice with minimal preparation. Clearly, the relationships between the length and quality of a doctor's training and his clinical skills are not simple. Medical students who rank high in their class obtain "better" internships and residencies than lower ranking students; yet there are also doctors of great skill whose training was short and in an undistinguished institution. There are also, undoubtedly, doctors who obtain excellent training but do not become clinically skilled. There is little doubt, however, that for the great majority of physicians, longer training is associated with greater clinical competence. Perhaps this greater competence reflects the training—or perhaps the doctors who select longer training are potentially more competent (or differ in some other significant way) to begin with. For example, Peterson, Andrews, Spain, and Greenberg ( 2 3 ) , as a sequel to the study of North Carolina general practitioners, studied another group of doctors with long clinical training. They found that as hospital training was prolonged, the doctors as a group became better and the variation among them became less. Perhaps the increased competence reflects the longer training; yet it is equally possible that the doctors who selected themselves for each additional year of training were more and

WHY

STUDY

CAREER

DECISIONS?

3

more alike in certain respects, and potentially more skilled at the start. Length of training is not the only significant variable; place of training may be important. Clute found, for example, that the clinical skills of general practitioners in practice were not related to the amount of nonteaching-hospital training they had received but that these skills were definitely related to length of training in teaching hospitals. Again, this may be a function of the training in the two kinds of hospitals, or of the men who select (or are selected for) this training, or both. Thus, in studying the general question of the quality of physicians' performance, it becomes important to ask: W h o chooses longer training and who chooses less? Why? W h o chooses teaching-hospital training and who does not? Why? W h a t are the roles of personal variables—intellectual, financial, family background, and the like—in these choices? W h a t other factors influence them? Until recently, relatively little attention has been given these questions. In contrast, much attention and study have been given to the selection of medical students, a question of importance, of course, but one which should not be allowed to obscure the basic concern of medical education: the competence of the graduate physician. A few studies have been concerned with what happens to the students after they leave medical school, as exemplified by studies in which students were asked what type of practice they expected to enter. The many specialty boards, with their important power of certification, are further evidence of the importance attached to residency training. Nevertheless, the great variability in the available hospital training programs and the freedom with which the medical school grad-

4

THE

TRAINING

OF

GOOD

PHYSICIANS

uate can select among them—freedom which is not to be criticized as such—results in the unfortunate situation that many doctors obtain poor hospital training that gives them little help in becoming skilled doctors. It is, therefore, important that the choices that the doctor makes about his training and about his practice be examined. Such an examination must be made with full realization that a student's choices will be modified by his material and intellectual resources, as well as those of his family; furthermore, the complexity of the factors which influence these choices should not be neglected. The medical students may attend a public or a private medical school; one that is large or small; a school with a long tradition and a wealth of faculty or another that has recently been formed. Attendance at one type of medical school may lead easily into a hospital training program that is seldom open to the students of other schools. In this study the medical student, the school, and the hospital have been examined to see how each is associated with the doctors' preparation for practice and selection of a field of practice. Naturally, there was interest in finding any facts that might be of predictive value. Is there, for example, any type of medical school student who is more likely than any other to prepare himself well for practice? Such student characteristics were not the only subject of interest. It was hoped that it would be possible to determine whether the circumstances which limit the preparation of some doctors are remediable. Good clinical training is not the only determinant of good medical care. Medical care insurance, hospitals, and other facilities are also necessary. However, competent doctors are the most essential single element. Since the measurement of a doctor's competence requires difficult and exhaustive direct studies of

WHY

STUDY

CAREER

DECISIONS?

5

practice and is therefore not practical on a large scale, this study has concerned itself with the doctor's hospital training. Although training is not as good a measure of the outcome of the medical education process as is competence, the evidence from direct studies that training and competence are strongly related justifies our use of training as the critical characteristic under investigation.

II / Study Methods and Approach T H E P U R P O S E of this study is to determine the factors which influence some doctors to obtain longer and better training and others to obtain shorter and poorer preparation for practice. Since certain types of practice, such as neurosurgery, are almost invariably associated with long periods of training in teaching hospitals, whereas other fields are associated with extremely variable periods and types of hospital training, the question "training for what?" is implicitly part of the study purpose. T h e approach we have taken is to consider first the present professional and family circumstances of the doctors included in the study. Next we examine the respondents' social and economic backgrounds, and relate these to the choice of a field of specialty practice. T h e respondents' academic performance in medical school and their subsequent hospital training are then related to their social and economic backgrounds and circumstances in medical school. Finally, we consider how the respondents evaluate the medical education and hospital training they received. This evaluation includes: whether they received as much training or the type of training they now believe they needed, and if not, why not; what shortcomings they now perceive in the educational process they experienced, and what they feel should be done about these deficiencies; and in retrospect, how they would restructure their training experience if they had a chance to do it over. Selection of Population

Sample

Most empirical studies of the training and career decisions of doctors have been based on samples of medical students. T h e

STUDY

METHODS

AND

APPROACH

7

student has been asked to indicate what additional training he expects to receive and what practice relationships he expects to enter into after graduation. (2, 13, 19) The difficulty with this approach is that many medical students may not yet have made decisions about these questions, and even those who have may well change their minds as the result of circumstances they encounter during their hospital training. For this reason, we decided that it would be more useful to approach doctors who had already completed their hospital training and had made at least tentative decisions about their specialty and practice relationships. The decision to study recent medical school graduates was made for several reasons. The doctors who attended medical school during World War II did so under very unusual circumstances that have little relevance to what might be regarded as normal. On the other hand, the postwar period saw a vast expansion in the available internship and residency programs and a marked increase in inclination among doctors to pursue extensive training before going into practice. The experiences of the doctors who finished medical school before the United States entered World War II are, therefore, not relevant to the problems of today. In questioning doctors about events that occurred during their internship and residency, it is clearly desirable to tax memory as little as possible; this provided another justification for the limitation of the study to men who had graduated recently. Since planning for the study began in 1960, we decided to include 1954 graduates, assuming that almost all of these men would be in practice seven years after the event. It was expected that men graduating in this year would be more representative of normal medical school output than in the earlier postwar period when a large proportion of medical students were veterans and, therefore, both older and recipients of

8

THE

TRAINING

OF

GOOD

PHYSICIANS

support under the G.I. Bill. The 1950 classes were selected as the second group of study for several reasons. Since they were removed by four years in time from the 1954 class, they might be sufficiently remote and sufficiently different to show any trends that were developing with respect to hospital training and distribution between different fields of practice. Examination of this group was expected to yield information on the extent to which financial aid to veterans affected either the length or the type of hospital training pursued. All the graduates of the classes of 1950 and 1954 in a limited number of medical schools were studied instead of nationwide random sampling for several reasons. The group of students who go through a given medical school together have a common experience. Because of this uniformity of educational experience individual student or family characteristics should have greater meaning. Furthermore, any possible influence of students upon one another can be measured best under circumstances where information on entire classes is available. Finally, this method permits examination of the school itself (or type of school) as a variable with possible influence on students' choices. Selection

of

Schools

The medical schools studied were selected on the basis of characteristics important to the goals of this investigation; varied data contributed to this decision. There is ample evidence from studies of Dickenson and others (10, 29) that the graduates of publicly supported and privately supported medical schools differ in their tendency to enter general or specialty practice or teaching and research. Public and private schools also have different problems in selecting medical students. Moreover, within the publicly supported and privately supported groups certain

STUDY

METHODS

AND

9

APPROACH

schools have, for years, tended to produce more teachers and research workers, others have tended to produce specialists, and still others to produce chiefly general practitioners. Schools of all three types were included in the sample. Thirdly, it was felt that the size of the medical school might have an influence upon students and hence upon their subsequent training. Lastly, it was reasoned that geographic location of medical schools could not be ignored even though there is no evidence that location, per se, has an effect upon a medical school's output. Church related and predominantly Negro medical schools were excluded because they would have expanded the study beyond a size judged to be feasible. Although both of these groups are small, giving them adequate representation would have necessitated unduly heavy sampling within each category. Another consideration in selecting the schools was the possibility of comparisons with the longitudinal studies being conducted by the Association of American Medical Colleges (AAMC), which has been following the careers of graduates from twenty-eight medical schools since 1956. The institutions which are the subject of this report are mostly included in the AAMC study; a few are not. Twelve schools were selected to provide a balance of the characteristics enumerated above. The number of schools by categories for each of the four characteristics are as follows. Ownership Public Private Traditional output General practitioner Specialty Teaching and research

Number of schools

6 6

Size Large Small

Number of schools 5 7

Location 5 4 3

East Midwest South West

4 3

2

10

THE

TRAINING

OF

GOOD

PHYSICIANS

A total interview sample of about 2000 doctors was estimated as necessary to permit breakdowns by various types of practice, school, and other characteristics. A study of this size was also deemed to be feasible. T h e 1950 and 1954 graduates of the twelve selected medical schools totaled 1887, a reasonably close approximation of the sample size desired. Questionnaire Design and

Administration

T o make feasible a study involving doctors from twelve medical schools from all parts of the United States, a mailed questionnaire was employed. Since doctors are characteristically very busy people and a considerable amount of information was necessary from each respondent, a questionnaire was developed that was thorough but still could be completed in a reasonable length of time. W h e r e possible, questions were phrased so that they could be answered by a check mark; other questions were phrased in such a way that the doctor could answer with one or two words. T h e questionnaires were pretested on a number of practitioners to eliminate, so far as possible, ambiguity or bias in wording.* T o insure a high response rate, each of the twelve schools selected for inclusion in the study was visited by a member of the study staff. T h e purpose and nature of the study was explained and each school's cooperation and support was elicited. T h e dean of each school agreed to send a letter to all of his 1950 and 1954 graduates urging them to cooperate in responding to the questionnaire. Assiduous follow-up of nonrespondents resulted in the ultimate return of questionnaires by 1771 doctors—an overall 94 * A copy of the questionnaire may be obtained from the authors.

STUDY

METHODS

AND

11

APPROACH

percent response rate. Considerable data was available on the nonresponders and is included in many of the tables of this report. For example, all of the nonresponders could be categorized by class rank, and the composite Medical College Admission Test (MCAT) scores were known for most of the 1954 graduates in the nonresponder group. Further, the place and type of practice could be determined for some. The small size of the nonresponder group precludes any substantial bias, due to lack of response, in the general findings of the study. Most of the nonresponders simply chose not to fill in the questionnaire. A few did not receive the questionnaire because they could not be reached despite an extended search for their current address and location. The available information does not indicate any systematic selection of nonresponders. The percentage of graduates from each school who returned the questionnaires are as follows. Public Schools A B

C D E F All Public Schools Private

Schools A' B'

1950 86 98 84 86 93 88 89

1954 95 94 94 98 93 98 95

95

95 94

D' E' F' All Private Schools

91 93 91 98 93

92 98 100 99 95 96 97

All

91

96

C'

Schools

86

Combined

93

12

THE

TRAINING

OF

GOOD

PHYSICIANS

The numbers of respondents in the four groups were as follows.

Six Public Schools Six Private Schools Total

1950 334 430

1954 498 509

Total 832 939 1,771

The returned questionnaires constituted 15 and 14 percent of all graduates of United States medical schools in 1950 and 1954 respectively. A study of the returned questionnaires showed that they were completed with commendable care. The frequent addition of explanatory notes and very occasional criticisms of the questionnaire technique itself were taken as further evidence of the conscientiousness of the doctors in providing information. Data

Analysis

Coded responses to all questions in the questionnaire were transferred to punched cards and answers were tabulated by private or public schools and by year of graduation. Tables were prepared to show how the responses on each training and career decision variable were related to the responses given to every other pertinent question in the questionnaire. Most data reported in this volume will be presented in terms of such contingency tables. This was done so that any characteristic that proved to differentiate the schools in terms of their graduates' decisions about training or careers could be held constant for purpose of analysis. Two questions attempted to elicit the "value orientation" of each respondent, that is, the criteria he felt to be most important in identifying a good doctor or a good medical student. The scale originated by James Coleman, Elihu Katz, and Herbert

STUDY

METHODS

AND

13

APPROACH

Menzel(8) was utilized, adjusting the wording to fit the context of this study. Analysis of variance was conducted to relate various school characteristics ( for example, public or private, size ) to (1) traditional output (graduates' distribution by fields of practice—the percentage distribution of general practitioners, specialists, and teachers and research workers); and (2) the various characteristics of the training obtained by graduates—length of residency, type of residency hospital, and so forth. The size and location of the medical school was found to have no measurable effect upon its graduates' hospital training or selection of a field of practice. Greater variance was found between the public and private medical school graduates in both 1950 and 1954 than was found within either the public or private groups. As the accompanying tabulations show, the difference between the public and private school graduates was marked and consistent when examined by proportion of graduates in general practice and length of residency. (P usually = s "Ê

eu (U C

O

ü

os oo es CO

'•uS §S .¡S y o 5 « . a o cu e CM fe00 c/5

js

DECISIONS

ABOUT

FIELD

OF

PRACTICE

51

Intellectual interests were stated to be important more often by the private school respondents than by those from the public schools. While 81 percent and 87 percent of private school graduates in 1950 and 1954 said they were much influenced by this factor, only 72 percent and 74 percent of the public school graduates of these years responded similarly. This difference is not significant for 1950, but is for 1954 (CR 1.7 and 3.5). Interest in certain types of patients and financial circumstances were more important in the public than the private school groups. This was especially true of financial circumstances —29 percent and 23 percent of the 1950 and 1954 public school respondents stated that finances were important, as compared with 18 percent and 11 percent of the 1950 and 1954 private school respondents, and these differences are significant (CR 2.4 and 3.5). If doctors are grouped according to the field of practice they reported in 1961, it is apparent that quite different combinations of factors influenced the choices of each group. General practice. Doctors who had chosen general practice were less influenced than all other respondents by intellectual interests, and more influenced by their interest in patients, financial circumstances, social pressures, and spouses' expectations and needs; these differences are significant. Only 57 to 60 percent of the general practitioners indicated that intellectual interests were important, compared to 84 to 90 percent of all other respondents; this difference is significant, both in public and private schools, in both 1950 and 1954 (CR 4.9 and 7.3 in public schools, 4.1 and 4.1 in private schools, in 1950 and 1954). Conversely, general practitioners listed financial circumstances as important influences on field-of-practice choice much more often than did other doctors; in 1950, 59 percent of the public school

52

THE

TRAINING

OF

GOOD

PHYSICIANS

and 52 percent of the private school general practitioners reported this influence, as compared to only 17 percent of all other public school graduates and 15 percent of all other private school graduates (CR for difference is 6.9 for public school graduates, 4.0 for private school graduates). The greater importance to general practitioners of financial circumstances was equally significant in 1954 (CR 5.5 in public schools, 6.7 in private schools). Similarly, interest in certain types of patients was reported to be an important influence significantly more often by general practitioners than other respondents in both public and private schools in 1950 (CR 4.3 and 4.2) and in public schools in 1954 (CR 4.1), with more than half the general practitioners listing this factor as compared to less than a third of all other doctors. The percentage of general practitioners reporting social pressures as an important influence on career choice ranged from 25 to 37, while only 8 to 14 percent of all other doctors listed social pressures (CR 5.0 and 3.9 in public schools, 2.6 and 3.4 in private schools, 1950 and 1954). Finally, general practitioners were significantly more likely to list spouse's expectations or needs as an important influence. In 1950, 41 percent of the public school general practitioners did so (versus only 12 percent of all other public school graduates) and 31 percent of private school general practitioners did so (versus only 11 percent of all other private school graduates); similar differences between general practitioners and all other doctors were found in 1954 (CR 5.9 and 5.1 in public schools, 1.9 and 2.5 in private schools, 1950 and 1954). In summary, then, general practitioners were significantly different from all other doctors in the reported influences on their choice of a field of practice—more influenced by finances, social pressures, their spouses, and interest in certain types of patients, and less influenced by intellectual interests. These characteristics

DECISIONS

ABOUT

FIELD

OF

PRACTICE

53

overrode any differences between public and private school graduates: among all general practitioners no significant difference between public and private school graduates was found. Internal medicine. Intellectual interest had a greater influence on internists' choice of specialty than on the choices of any other group of respondents with the single exception of teachers and research workers. From 92 to 98 percent of internists in public and private schools, and in both 1950 and 1954, indicated they were influenced by this factor. In comparison, this influence was reported by 72 to 87 percent of all other doctors; the difference between internists and others was significant (CR range 4.1 to 6.0) except among private school graduates in 1954. Its importance is further highlighted by the fact that fewer internists (as compared to all other doctors) indicated they were influenced by special interest in patients, financial circumstances, social pressures, or spouses' expectations. Surgery. The percent of surgeons influenced by intellectual interests in the choice of their specialty was about the same as that for all respondents in the study. As compared with all other doctors, however, significantly fewer surgeons were influenced by special interest in patients. Only 16 to 20 percent reported this influence, compared to 30 to 42 percent of nonsurgeons (CR 4.6 and 5.7 in public schools, 3.9 and 2.2 in private schools, 1950 and 1954). No marked differences between surgeons and nonsurgeons appeared with regard to financial circumstances, social pressures, or spouse's expectations. Teaching and research. The graduates who were teachers and research workers overwhelmingly reported that they were influenced by intellectual interests in selecting their careers. In the private schools nearly 100 percent so reported; the corresponding proportions in the public schools were 84 percent of the 1950 and 92 percent of the 1954 classes.

54

THE

TRAINING

OF

GOOD

PHYSICIANS

This emphasis on intellectual interests was especially marked when teachers and researchers were compared with all other doctors (CR 2.1 and 5.0 in public schools, and 6.5 and 4.5 in private schools, 1950 and 1954). Financial circumstances, social pressures, and spouse's expectations were of little importance in making the decision to follow a teaching or research career. Less than 10 percent of teacher-researchers indicated they were much influenced by any of these considerations. Other specialties. The pediatricians were the only group in which intellectual interests were not the most frequently reported consideration in choice of a career. Except among the 1954 public school graduates, interest in certain types of patients was reported to be important more frequently than intellectual interests. W h e n pediatricians were compared with all other physicians, this emphasis on interest in patients was significant (CR 8.6 and 6.1 in public schools, and 6.9 and 12.0 in private schools, 1950 and 1954). In general, the responses of obstetrician-gynecologists were similar to those of pediatricians, especially in interest in patients. As compared with all other physicians, obstetrician-gynecologists from the private schools (but not the public schools) in both 1950 and 1954 reported patient interest significantly more frequently as an important influence on career choice.

Father's Background

and

Education

The education of the father of a medical student might have both direct or indirect effects upon the student's career and subsequent choice of a field of practice. It is known that educa-

DECISIONS

ABOUT

FIELD

OF

PRACTICE

55

tion and income are related, with larger incomes being associated with more education. Thus, the influence of education might really be an indirect one relating to family finances. A substantial number of medical students (approximately one in eight in the present study) have fathers who are doctors; as we will see later, these students are among those with the longest residency training. Either finances or family educational tradition might explain this. About half of the doctors reported that their fathers had obtained a high school education or less (Table 17). About a fourth of the doctors had fathers who had gone to college or graduated from college, and another fourth had fathers with postgraduate college educations, including, of course, the fathers who were doctors. There was only one significant difference between the public and private school graduates' descriptions of their fathers' education; in 1954 more fathers of the private school graduates had done postgraduate work ( C R 2.6). T h e data presented in Table 15 suggested that teachers and researchers, in general, had fathers with more education, and that general practitioners, as a rule, had fathers with less education than was the case with other physicians. T o examine this possibility more clearly, the data were regrouped to permit comparison of teachers and researchers with all others and of general practitioners with all others, with respect to father's education. These comparisons yielded rather few significant findings. Among general practitioners, for example, significantly fewer 1954 graduates of the public schools had fathers with postgraduate educations. Among teachers and researchers the tendency to have fewer fathers with no college training was significant in one class (private schools, 1950), as was the higher proportion of fathers with postgraduate training (public schools, 1954).

ti S < -S o

rt O -O e ctf O O -tí

o V> u ni Ul CM M-i O

o 2 « i» tí o •i-i +•» rt O 13 u

r^ O «

T3 •ferH

H

u o -O O c o1-H ••M

H

v

-tí o kl rt ^O "2 tí rt c« o u H



o oc •M e u o K a> 0-, . r1-I w i-J «

8 T3

< -tí

i-, u -tí

a i*t «u u

o


a

H

feb

< rt

rt •ti 5

« 3



oo

en

CM

oo m

O o -tí u

O

« o

-d o

o

rt U

rt u

'S IU g

-d

V

E bO d

u

jc ¿

U * O •J

tí c4

(O I s H

rH

VO O vO

a!

-2 "o a> o o

a, D

tn \o N m h 1/1 N N N H H

m rH m

rH CS rH rH

o s a. -o -a

tf) h o •M u o -ä Vu O ä o •M

Ih U *

O

rH

rH

r-i

rHi-HCNTfCSCSt^CI rH rH rH rH

hJ

oo m t^ o «o H

PO oí O N

io

vomvooovouimov ^ rH

t2u u (50 nt C uu I* ü a, fO o w S

U t4 O 2u Ü4

< -H

O

O I N

I—I T-H

I O CO

CO CO

ON

C

« O

£ 2 CL,

3

U2

NO

ON

CO EN

CS

ON

O

°

NI

_ E

u - O C RT

§ J S U V « O

2

2

0

Y

- E

1 3

"13

. Y - D ^ S

. A T 3 — E ?

OH O

OH

c/3

_yfi

S

•a 5 c T) rt

-

ov

Ov

cs •t

CS CO -O fi a,

o

VO CO

cs

Ov

O

CS

CM

co

CS

co

CO

Y-H

Tj- -h

CS

VO

i

co th

i-H oo t^ m vo m rf

i1

1i

m co

m cs

vo

vo

tj-

cs T-i

m vo

*—i vo

m *h

m

CO

i-H m Tt

-t

CO vO

cs

O in

cs

m CO

vo

i

oo m CO H

m ov

o

co

o

CO

vo o m

m i r »

in

i-i

1

i

rUO C U e -C u (t r3 3 •18» op O C Ld s ij2 4>

CM in eg t--

bJD C

.rtu .3 •M

ON IT)

ra

m ON

^ .y 3 «

tuo c

NO en

O e¿

U C3U o in ON

15 ñ •g60 ^y *cL §> U 4> 3 w s ««

NO iH 00 04 cu >> . 4-»

00 « ^ .£
-tí

2 o *-> O

O I-

-a ••— -tí rt o S

CL,

vO w hJ M

< H

c o ex, V bD


O -o c rt t»0 c • w-i M O rt u-* 4 l-l O

m io es

O

vi i-i 0

V 6

o

CS

¡50 c u ni u Ui V J« 4-» O

O ID ON — i1

00

•51-

CN

V 4 n->i bD *t-i u c Oh 0 rt su s rt 1->» 4

Ui 0) JuS o O 3 3 Û4 o 'S1 2

1-4 O V 2

W O u m I 3 0 VI

OV U-)

cs

CS CS CS

m

CS

Ù0 c o rt 4) •M

U">

tN CO

ID (O

60 S -G U *

\o

so

•o e ON ON •t

Ol 0> ni u S«

ON o\ t

m

O O m

a

O O m

o

CL,

O

z

HOSPITAL

TRAINING

DECISIONS

155

public school group (CR 1.98). The absence of a similar difference for the 1950 graduates may be due to the G.I. Bill which was a frequent source of support reported by the members of that cohort. Similarly, lower earnings are associated with the choice of a major teaching hospital for residency except among the 1950 private school graduates (Table 69). Although this tendency is quite consistent, the associations are not strong. The frequency of reported no residency appears from Table 69 to be consistently associated with earnings during medical school. Among the differences between residency and no residency that were significant are those for 1950 public school graduates with earnings of $499 or less (CR 2.2), the 1950 private school graduates in the same income category (CR 1.9), and the 1950 public school graduates with the highest earnings (CR 2.4). None of the differences for the 1954 graduates were significant. The fact that none of the differences found for the private school groups were definitely significant may again be due to the small size of the group without residency training—37, or 8 percent of the total, in 1950 and 21, or 5 percent of the total group, in 1954. Among doctors with a residency, the length of residency training was also associated with the history of earnings in medical school given by public school graduates (Table 70). Testing of the tendency of public school graduates with high earnings to take shorter residencies was significant for the 1950 graduates (CR 2.7) but not for those of 1954. While this tendency was also present among the private school graduates, it was neither very strong nor regular and was not significant. The results of this closer examination of the influence of finances on training are consistent with those presented in the

u

o 13

00

en

cn

°

vO

«s

nin

VO

m CO

a

« V 00

"2 "S V t-1

"t

CS

I

'S o rt ^ 41 >

O

00

Î

0

o O")

C

5 6 c«

in Ul a. rt i> 00 >-. o y-H T3 C M rt « "o O < - tOí H CA

00